Exploit found which can lead cheating

This is what I’m talking about:


-edit- removed some screenshots

2 Likes

Good luck spotting it with height.

2 Likes

Simple answers

Is Cheating in group rides Cheating YES
Can you use weight/height explot if its not a race NO
Is weight/height manipulation ever considered a feature NO

3 Likes

You can always cheat yourself. Wonderful thing, self-delusion.

3 Likes

Yes also check Alpe dZ segment best times this past week. 4 or 5 stand out as managing quick times riding 1 or 2 Wkg below anyone else around them.

A couple of days ago, I posted my “first thoughts” about this (post 340) and said I hadn’t decided whether to post my “second thoughts”. What held me back was the timing – I didn’t want to post those thoughts within 24 hours of the world championships, as they related to the credibility of events like this and I’ve got a lot of respect for the competitors.

Thankfully, since then, @Eric Min has responded and his statement included measures to cover off the key concern I had. This is what I wrote but didn’t publish on Friday:

My second thoughts:

Having recently typed up some anti-doping guidelines for a sports governing body, I went looking for Zwift’s guidelines about how to report issues. But the only reporting process I could find in Zwift’s esports guidelines tell you how to report people who cheat - not gaps in the software.

I think it’s understandable that Zwift didn’t add this to their guidelines earlier - this is a new sport where you couldn’t predict everything which would come up. But for the credibility of esport going forward, I feel there has to be a clear way of reporting software issues which can be exploited, and a process which will be followed so ways to cheat are stopped quickly - not “longer term”. I think this is important for the credibility of all online sport, of platforms like Zwift, and of governing bodies which associate with it.

For the sake of credibility, I think we need a clear process of how we can (privately) report ways of cheating with the software, and clear details of how they will be handled, so when opportunities to cheat appear, we can have the confidence to see them quickly get sorted.

…And this is how Eric Min covered this off in his message:

The suggestion of a “bounty” gives me some confidence that the type of system Eric is talking about should have the tracking and transparency this type of thing needs to be credible. This is great, but I have to say my confidence is tempered by Zwift’s track record of fixing things “in due course” or “Zoon”. It’s this which is making me think “it’s time for me to think of Zwift as more like ‘just a game’ and less something to be so passionate about.

I’d like to thank great Zwift staff like @xflintx and @James_Zwift and @shooj for responding to things quickly on here, and I’ve loved the results of the work which has gone into recent improvements to racing with cat enforcement. Thanks to everyone else as well for the engaging conversations we’ve had about detailed topics.

But the last few days have made me want to step away from the madness on here. I’ll still be a Zwifter, I’ll still push like heck in the Tour of Watopia, but I’ll be quieter.

So have fun everyone, and enjoy going round and round in circles - whether you’re racing in crit city or talking about race ranking. :wink:

Ride On.

3 Likes

I couldn’t see how to check Alpe segment attempts over the past week but it appears there is a new KOM who averaged 19.4 mph and did the climb in 27 min.

I’m sure he’s wondering why he got kicked up to Cat B.

I guess I need to do more intervals.

1 Like

Good luck with that too.

Zwift still haven’t resolved the problem with segments not being recorded for large numbers of people if the event finishes on the same end point.

And it was for some.of rapha stage 1.

I’ve raced the alpe in TTT, FRR and now Rapha and still have no segment time at ZP.

1 Like

Yeah, I’ve been up that stupid thing 12 times or so and don’t have a time on ZP. Frickin’ annoying…

2 Likes

Just did some casual browsing of the leg I did of rapha today.

Went into sprints and kom tab where everyone’s watts and w/kg is listed for each segment.

Picked 4 at random, 2 guys had a consistent weight (within 0.5kg) for the 3 segments, 2 guys had a variation in weight > 10kg. (Although one guy was lighter for the sprint and heavier for both climbs)

I may be missing something but isn’t it a simple maths calculation? The database should be able to do that for segments in a race as it is processing and flag any variation > 5kg across segments in the same race ??

I get that its avg W for a segment, but even so, the kg end of the equation doesn’t change.

I’m am willing to accept I may have done this wrong and look stupid though. @James_Zwift

That 2.6 w/kg doesn’t mean much, especially if it’s a flat race where speed is determined by watts more so than w/kg. Heavier riders will always have an advantage if they’re producing the same or lower w/kg and more watts.

They could have also been drafting off riders with a higher w/kg most of the race and have a better 5 min, 1 min power, and sprint. That screenshot you provided doesn’t give enough detail to reach the conclusion that the 2.6 w/kg rider is cheating.

2 Likes

Before the thread gets locked for people going off on a witch hunt:

Zwifts API does not record average power for a segment accurately for all riders all the time. You cannot use that data on ZwiftPower to determine if someone cheats.

5 Likes

Yeah, but that was probably the Road to Sky stage. There were a coupla “C” riders that finished second and third in my event that put down 2.3 and 2.4 whilst weighing 70+kg. I’m reasonably sure they were cheating. There have a few in every single one of the Rapha events. It’s pretty easy to see because there was so much climbing. Really no where to hide in those events.

2 Likes

I have to do around 260w average if I want a fast time up Alpe, even though I’m light.

I can’t imagine doing 2.3 or similar even for a while ride, I’d be slow.

1 Like

I am utterly appalled that this has been known for so long.

I appreciate hearing that the CEO acknowledged this debacle, but the way this was handled showed its about profits. The racers are hardcore users, so no need to worry about their experience, just the newbies on trial gotta sign up. Give em another badge or something!

It bothered me enough that I have thought about going back to FulGaz. Messed me up for a few days.

However, I would rather make a stink to make change. Yet, with a private company, I have no rights other than where I spend my money and leaving posts on forums that I am despondent. Again.

2 Likes

The ride was on ventoux… (or atleast the ride before was and they used the hack on that ride to)

Plus there are screen shots where you can see the watts and wkg… one image shows 90kg rider, the next a 70kg rider.

1 Like

Can I check are you looking at fastest segment times or ZPoints. Alpe not included in ZPoints.

I’ve been up Alpe around 12 times and have 8 times on my record but know not all group rides are included.

Have you done
Segments
Alpe
Me

I don’t see how zwift can on the one hand say the onus is entirely on us to identify potential cheats and bring them to zwift’s attention, and on the other hand say we are not allowed to discuss dodgy results.

At least, not if we are still clinging to the fantasy that this fiasco is primarily to do with maintaining race integrity rather than being a public relations exercise.

I feel that soon they’re going to need a bigger carpet.

6 Likes

Rider still not been DQ who has used the bug to improve his race time.

A) Zwift are aware as per Eric’s comments and have chosen to leave that user in the GC standings against those using the system as intended.
B) Zwift have no idea who is and who isnt using the bug and want the community to report it as a token effort.

Which do we think is the plausible?

There may well be glitches in power calculations I don’t know and probably difficult to investigate.

So are there problems with times?

I think people here are just trying to demonstrate to Zwift possible indicators which might suggest further investigation is warranted.

I suspect due to recent publicity about the whole weight doping affair that several riders have tried this over the past week. I am prepared to give them the benefit of doubt and hope, possibly naively, they will not need to experiment again.

I don’t know about others but I’m slightly disappointed that Eric/Zwift think I want a ‘bounty’ for helping identify bugs/fixes - I don’t. I want to help provide a more honest better racing experience, that is reward enough.

To me providing checks Zwift might consider making is one thing but calling out cheats is a far more difficult and unpleasant practice which should ideally be identified and dealt with by Zwift.

1 Like